Colorectal cancer is one of the three highest incident cancers in this country with 140,000 new cases and 52,000 deaths annually. Evidence strongly suggests that colorectal adenomas are the premalignant stage of this disease. Removal of adenomas, the search for synchronous adenomas, and periodic surveillance for metachronous adenomas, the search for synchronous adenomas, and periodic surveillance for metachronous adenomas could reduce the incidence and mortality of colorectal cancer. Identification of individuals with adenomas, especially those with the more premalignant types with a high probability of recurrence and development of colorectal cancer, would be critical in planning cost-effective control strategies targeted to a high risk subset of the population. Familial risk factors have assumed importance recently as a basis for identifying such high risk individuals. Defined genetic syndromes (Familial Polyposis, Family Cancer Syndrome) account for only 2000-7000 new cases annually. Recent evidence suggests that a significant proportion of the sporadic cases also may have a genetic basis. Study of families of sporadic cases of colorectal cancer as well as families of probands with sporadic adenomas could help clarify both familial risks and the genetic basis for sporadic adenomas and adenocarcinoma of the colon. These are the overall objectives of this proposed study. Our hypothesis is that familial factors confer a significant risk for colorectal adenomatous disease and colorectal cancer and that this spectrum has a genetic basis. These studies will be based on the National Polyp Study, a major resource of well characterized patients with normal colons, all adenoma stages, and colorectal cancer. The study will have two major approaches: a) to study the family history of colorectal cancer in NPS probands by verified questionnaire; and b) to study families of NPS probands by flexible sigmoidoscopy to identify members with adenomas. A risk index relating family history to adenoma pathology will be developed and utilized to predict adenoma recurrence. The relationship between adenomas and colorectal cancer will be validated by comparison of the family history in probands with and without adenomas. Risk of adenomas in families will be determined directly by sigmoidoscopy and the hypothesis that this risk is controlled by a dominant gene will be tested. Nutritional differences among families will be assessed as a control variable. Clarification of familial risk factors and the genetic basis for adenomas and colorectal cancer could result in improved approaches for primary and secondary prevention. A smaller proportion of the population could be targeted based on increased risk for greater cost-effectiveness and compliance.